Large-scale synthesis modules have been developed and used for the preparation of a number of radiopharmaceutical compounds, including 2-deoxy-2-[F-18]-fluoro-D-glucose (FDG) and 3′-deoxy-3′-[F-18]-fluorothymidine (FLT). Such modules or reactors occupy a large amount of space and the chemical process requires longer reaction time cycles than desired for the preparation of the labeled compounds. These modules and reactors are also difficult to modify for the research and development of new compounds and probes. But their main drawback is that the reactions take place with reduced efficiency arising from tremendous dilution of reagents necessary only for macroscopic liquid handling.
An earlier microfluidic chip has been developed by Tseng, et al. from the Department of Molecular and Medical Pharmacology, UCLA. This microfluidic chip is capable of producing [F-18]FDG on a 58 microcurie (μCi) scale. However, the design and configuration of this microfluidic chip have certain limitations, and the chip does not permit the preparation of the labeled product on a significantly larger scale.
The synthesis of the [F-18]-labeled molecular probe, 2-deoxy-2-[F-18]-fluoro-D-glucose (FDG) is based on three major sequential synthetic processes: (i) Concentration of the dilute [F-18]fluoride solution (1-10 ppm) that is obtained from the bombardment of [O-18] water in a cyclotron; (ii) [F-18]fluoride substitution of the mannose triflate precursor; and (iii) acidic hydrolysis of the fluorinated intermediate. Presently, FDG is produced on a routine basis in a processing time (or cycle time) of about 50 minutes using expensive (e.g., >$100K) macroscopic commercial synthesizers. These synthesizers consist, in part, of an HPLC pump, mechanical valves, glass-based reaction chambers and ion-exchange columns. The physical size of these units is approximately 80 cm×40 cm×60 cm.
Inevitably, a considerable decrease in the radiochemical yields of the resulting probe are obtained from these commercial synthesizers because of the long processing times, low reagent concentrations and the short half-life of [F-18]fluorine (t1/2=109.7 min). Moreover, because the commercialized automation system is constructed for macroscopic synthesis, the process requires the consumption of large amount of valuable reagents (e.g. mannose triflate), which is inefficient and wasteful for performing research at the smaller scale. For example, the required radioactivity for FDG PET imaging of a single patient is about 20 mCi, which corresponds to about 240 ng of FDG. However, for small animal imaging applications, such as for a mouse, only about 200 μCi or less of FDG is required.
Accordingly, there is a need to develop smaller or miniaturized systems and devices that are capable of processing such small quantities of molecular probes. In addition, there is a need for such systems that are capable of expediting chemical processing to reduce the overall processing or cycle times, simplifying the chemical processing procedures, and at the same time, provide the flexibility to produce a wide range of probes, biomarkers and labeled drugs or drug analogs, inexpensively. These miniaturized devices may employ polymers, such as PDMS-like elastomers that are inert under the reaction conditions.
Commercial large-scale synthesizers (e.g. Explora and CPCU) are capable of preparing up to 50 doses in a lab-sized operation. On a smaller scale, a microfluidic chip has been disclosed by Tseng, et. al. at the University of California, Los Angeles. The microfluidic chip has been demonstrated to produce 58 microcuries of FDG in a single run. However, the design of this microfluidic chip is such that it is not capable of scaling up by over a 1500 fold that is required to achieve a desired 100 mCi level of activity. In addition, the particular design of the reaction process does not permit a significant increase in the output or the yield.
In addition to the inability to scale up the UCLA microfluidic chip, the inherent design of the chip also limits the loading of reagent activity thereby limiting the reaction throughput. That is, the microfluidic chip requires over 1 hour to load a minimal (500 microcuries) activity onto the exchange resin, which is an unacceptable period of processing time given the short half-life of F-18.
As disclosed in the present application, the design of the microfluidic device overcome this throughput limitation in addition to a number of other advantages. In particular, the device is capable of producing the desired amount of radioactivity in a short (5 minutes) period of time, and the design of the device does not have internal factors limiting either parameters.